Ensuring a given transactional unit of work arrives at an appropriate server instance

Abstract

Ensuring that a given transactional unit of work arrives at an appropriate server instance. A server instance receives a request that is associated with a unit of work. The server instance that received the request determines whether it is responsible for the unit of work associated with that request. When the server instance is responsible for the unit of work, the server instance processes the request. When the server instance is not responsible for the unit of work, it attempts to register its interest in the unit of work. If this attempt is successful, then the server instance can process the request. If the attempt is unsuccessful, then the responsible server instance is located and the request is forwarded to the responsible server instance.

What is claimed is:  
     
         1 . A method of managing requests within a computing environment, said computing environment comprising a plurality of server instances, and said method comprising: 
 determining, by a server instance of said plurality of server instances, whether said server instance is responsible for a unit of work associated with a request received by said server instance; and    processing said request, when said server instance is responsible for said unit of work.    
     
     
         2 . The method of  claim 1 , further comprising registering an interest of said server instance in said unit of work, wherein said server instance is responsible for said unit of work.  
     
     
         3 . The method of  claim 2 , wherein said registering comprises atomically registering said interest at a coupling facility coupled to said server instance.  
     
     
         4 . The method of  claim 1 , further comprising determining which server instance of said plurality of server instances is responsible for said unit of work, when said server instance receiving said request is not responsible.  
     
     
         5 . The method of  claim 4 , further comprising indicating to a requester of said request which server instance is responsible for said unit of work.  
     
     
         6 . The method of  claim 5 , wherein said indicating comprises using a location forwarding reply to indicate to said requester the responsible server instance.  
     
     
         7 . The method of  claim 6 , further comprising building an object reference to be included within said location forwarding reply, said object reference indicating a location of the responsible server instance.  
     
     
         8 . The method of  claim 1 , wherein at least a portion of said plurality of server instances are replicas of one another.  
     
     
         9 . The method of  claim 1 , wherein said unit of work is a transactional unit of work.  
     
     
         10 . The method of  claim 1 , wherein transient state associated with said unit of work is protected.  
     
     
         11 . A method of managing requests within a computing environment, said computing environment comprising a plurality of server instances, and said method comprising: 
 receiving, within said computing environment, a plurality of requests that are associated with a unit of work; and    ensuring that said plurality of requests are processed by one server instance of said plurality of server instances, wherein a temporary affinity to said one server instance is provided for requests under said unit of work.    
     
     
         12 . The method of  claim 11 , wherein at least a portion of said plurality of server instances are replicas of one another.  
     
     
         13 . The method of  claim 11 , wherein said ensuring comprises using a registration protocol to register interest of said one server in said unit of work and a location forwarding protocol to forward any requests under said unit of work received by other server instances to said one server instance.  
     
     
         14 . The method of  claim 11 , wherein said temporary affinity continues to be provided, without aborting said unit of work, even if a connection to said one server instance is reestablished after being broken.  
     
     
         15 . A system of managing requests within a computing environment, said computing environment comprising a plurality of server instances, and said system comprising: 
 means for determining, by a server instance of said plurality of server instances, whether said server instance is responsible for a unit of work associated with a request received by said server instance; and    means for processing said request, when said server instance is responsible for said unit of work.    
     
     
         16 . The system of  claim 15 , further comprising means for registering an interest of said server instance in said unit of work, wherein said server instance is responsible for said unit of work.  
     
     
         17 . The system of  claim 16 , wherein said means for registering comprises means for atomically registering said interest at a coupling facility coupled to said server instance.  
     
     
         18 . The system of  claim 15 , further comprising means for determining which server instance of said plurality of server instances is responsible for said unit of work, when said server instance receiving said request is not responsible.  
     
     
         19 . The system of  claim 18 , further comprising means for indicating to a requester of said request which server instance is responsible for said unit of work.  
     
     
         20 . The system of  claim 19 , wherein said means for indicating comprises using a location forwarding reply to indicate to said requester the responsible server instance.  
     
     
         21 . The system of  claim 20 , further comprising means for building an object reference to be included within said location forwarding reply, said object reference indicating a location of the responsible server instance.  
     
     
         22 . The system of  claim 15 , wherein at least a portion of said plurality of server instances are replicas of one another.  
     
     
         23 . The system of  claim 15 , wherein said unit of work is a transactional unit of work.  
     
     
         24 . The system of  claim 15 , wherein transient state associated with said unit of work is protected.  
     
     
         25 . A system of managing requests within a computing environment, said computing environment comprising a plurality of server instances, and said system comprising: 
 means for receiving, within said computing environment, a plurality of requests that are associated with a unit of work; and    means for ensuring that said plurality of requests are processed by one server instance of said plurality of server instances, wherein a temporary affinity to said one server instance is provided for requests under said unit of work.    
     
     
         26 . The system of  claim 25 , wherein at least a portion of said plurality of server instances are replicas of one another.  
     
     
         27 . The system of  claim 25 , wherein said means for ensuring comprises using a registration protocol to register interest of said one server in said unit of work and a location forwarding protocol to forward any requests under said unit of work received by other server instances to said one server instance.  
     
     
         28 . The system of  claim 25 , wherein said temporary affinity continues to be provided, without aborting said unit of work, even if a connection to said one server instance is reestablished after being broken.  
     
     
         29 . A system of managing requests within a computing environment, said computing environment comprising a plurality of server instances, and said system comprising: 
 a server instance of said plurality of server instances being adapted to determine whether said server instance is responsible for a unit of work associated with a request received by said server instance; and    said server instance being adapted to process said request when said server instance is responsible for said unit of work.    
     
     
         30 . An article of manufacture, comprising: 
 at least one computer usable medium having computer readable program code means embodied therein for causing the managing of requests within a computing environment, said computing environment comprising a plurality of server instances, and the computer readable program code means in said article of manufacture comprising:    computer readable program code means for causing a computer to determine, by a server instance of said plurality of server instances, whether said server instance is responsible for a unit of work associated with a request received by said server instance; and    computer readable program code means for causing a computer to process said request, when said server instance is responsible for said unit of work.    
     
     
         31 . The article of manufacture of  claim 30 , further comprising computer readable program code means for causing a computer to register an interest of said server instance in said unit of work, wherein said server instance is responsible for said unit of work.  
     
     
         32 . The article of manufacture of  claim 31 , wherein said computer readable program code means for causing a computer to register comprises computer readable program code means for causing a computer to atomically register said interest at a coupling facility coupled to said server instance.  
     
     
         33 . The article of manufacture of  claim 33 , further comprising computer readable program code means for causing a computer to determine which server instance of said plurality of server instances is responsible for said unit of work, when said server instance receiving said request is not responsible.  
     
     
         34 . The article of manufacture of  claim 33 , further comprising computer readable program code means for causing a computer to indicate to a requester of said request which server instance is responsible for said unit of work.  
     
     
         35 . The article of manufacture of  claim 34 , wherein said computer readable program code means for causing a computer to indicate comprises using a location forwarding reply to indicate to said requester the responsible server instance.  
     
     
         36 . The article of manufacture of  claim 35 , further comprising computer readable program code means for causing a computer to build an object reference to be included within said location forwarding reply, said object reference indicating a location of the responsible server instance.  
     
     
         37 . The article of manufacture of  claim 30 , wherein at least a portion of said plurality of server instances are replicas of one another.  
     
     
         38 . The article of manufacture of  claim 30 , wherein said unit of work is a transactional unit of work.  
     
     
         39 . The article of manufacture of  claim 30 , wherein transient state associated with said unit of work is protected.  
     
     
         40 . At least one program storage device readable by a machine, tangibly embodying at least one program of instructions executable by the machine to perform a method of managing requests within a computing environment, said computing environment comprising a plurality of server instances, and said method comprising: 
 receiving, within said computing environment, a plurality of requests that are associated with a unit of work; and    ensuring that said plurality of requests are processed by one server instance of said plurality of server instances, wherein a temporary affinity to said one server instance is provided for requests under said unit of work.    
     
     
         41 . The at least one program storage device of  claim 40 , wherein at least a portion of said plurality of server instances are replicas of one another.  
     
     
         42 . The at least one program storage device of  claim 40 , wherein said ensuring comprises using a registration protocol to register interest of said one server in said unit of work and a location forwarding protocol to forward any requests under said unit of work received by other server instances to said one server instance.  
     
     
         43 . The at least one program storage device of  claim 40 , wherein said temporary affinity continues to be provided, without aborting said unit of work, even if a connection to said one server instance is reestablished after being broken.